CN109286640B - A communication method for adaptive adjustment based on satellite network status - Google Patents

Abstract

Translated fromChinese

本发明公开了一种基于卫星网络状态进行自适应调节的通信方法，本发明的系统收到用户的通信申请，系统同意请求后开始通信；通过获取通信卫星品质因数、通信带宽和用户发送功率以及通信链路长度和此次的传输波长等参数判断当前网络通信状态，将各个参数带入公式中进行计算，得到初始发送速率，并以此完成通信连接等操作；系统自适应调整环节，每完成k次数据包的传输，就重新计算丢包间隔C_i′并生成新的数据发送率X；本次通信结束。本方法可以根据网络实时状态做出自适应调整，方程式中的主要参数不需要预估设定，便于动态变化适应实际通信环境，更加灵活；相比于其他传输控制协议，本方法具有更好的适应性，以及更高的网络效率。

The invention discloses a communication method for self-adaptive adjustment based on satellite network status. The system of the invention receives a communication application from a user, and starts communication after the system agrees to the request; Parameters such as the length of the communication link and the current transmission wavelength determine the current network communication status, and bring each parameter into the formula for calculation to obtain the initial transmission rate, and use this to complete the communication connection and other operations; After k times of data packet transmission, the packet loss interval C_i ' is recalculated and a new data transmission rate X is generated; this communication ends. This method can make self-adaptive adjustments according to the real-time state of the network, and the main parameters in the equation do not need to be estimated and set, which is convenient for dynamic changes to adapt to the actual communication environment, and is more flexible; compared with other transmission control protocols, this method has better performance adaptability, and greater network efficiency.

Description

Communication method for self-adaptive adjustment based on satellite network state

Technical Field

The invention belongs to the technical field of wireless network communication, and relates to a multimedia service transmission control protocol in a satellite network.

Background

With the development of the times, the demand of people for internet multimedia services is gradually increased, and stream media applications such as IP telephony, video on demand, video conference, distance education and the like are greatly developed by means of a wireless network, wherein a satellite network communication mode is the future development direction. The satellite network has the advantages of flexible deployment, large coverage area and the like, and provides internet access service for a plurality of regions without cable network infrastructure. A Transmission Control Protocol (TCP) is a main data transmission protocol in a network, but since the TCP is a connection-oriented protocol, the real-time requirement of the service cannot be guaranteed, and the transmission efficiency in the aspect of multimedia data is low. While the UDP protocol oriented to non-connection will unfairly occupy a large amount of bandwidth when transmitting multimedia data service, greatly inhibiting throughput of other TCP streams, and aggravating congestion of the entire network, eventually leading to network crash, therefore, the Internet Engineering Task Force (IETF) proposes a transmission protocol based on rate — a TCP Friendly Rate Control (TFRC) to transmit multimedia service data. TFRC calculates the sending rate according to the TCP throughput equation and provides smooth data stream communication, the protocol maintains high friendliness for both TCP and UDP data streams, and its performance has been widely agreed in wired IP networks. However, the performance of the TFRC protocol in the wireless network is poor, packet loss caused by the high error rate of the wireless network itself and packet loss caused by router queue congestion coexist, and the receiving end attributes all packet loss to network congestion, so that the system cannot accurately find the congestion condition in the network, thereby blindly reducing the sending rate of data stream and causing the reduction of network performance.

At present, the extension and improvement of TFRC at home and abroad are mainly as follows: 1) TFRC wireless: the LDA algorithm (Loss differential algorithm) is added to the TFRC protocol. The protocol can distinguish lost data packets through an LDA algorithm, only network congestion packet loss is considered, and wireless loss packet loss is ignored, the current LDA algorithm mainly comprises Biaz, Spike, ZigZag and the like, the algorithms judge the loss type of the data packets according to delay jitter, namely packet loss classification is carried out through the relative one-way time (ROTT) of the data packets in a link. However, they all need to manually set a corresponding threshold, which cannot adapt to the environment of the wireless network, such as the transmission distance, and may encounter a high misclassification rate in some specific cases. 2) Multi frc: the core idea of the algorithm is to create multiple TFRC connections on the same link to make up for the deficiency of insufficient link utilization of a single connection. The algorithm is an end-to-end mechanism improvement and does not need any modification on network hardware equipment. However, the algorithm needs more resources to control the connection, and the data segmentation at the transmitting end and the data recombination at the receiving end also increase the system resource overhead, so that the comprehensive cost performance is not high. 3) MBTFRC: based on TFRC (measured-based TFRC, MBTFRC)3) MBTFRC, the data sending rate is adjusted by using passive bandwidth Measurement at a receiving end based on the measured TFRC (measured-based TFRC, MBTFRC), and the performance of a TFRC protocol is improved. The algorithm uses a window-based exponentially weighted moving average (WEWMA) to adjust the rate, and has certain flexibility and system stability. However, the algorithm needs to set two weight values, does not distinguish the type of network packet loss, does not consider long-time propagation delay, and is not suitable for a satellite network environment.

The above three algorithms are the main improvement ways of TFRC in the wireless network environment at present. However, these expansion methods are relatively fixed, the weight and the threshold value need to be estimated and set, and are not convenient for frequent change, and the wireless network environment is dynamically changed at all times, especially in the satellite network environment, and due to the influence of weather and communication distance in the area, the link state is good and bad, which cannot be summarized.

Aiming at the defects of the method, a TFRC protocol improvement algorithm TFRC-EA (environmental adaptation) for self-adaptive adjustment according to the channel state is provided. The method calculates the link error rate by acquiring the information such as the length, the wavelength and the like of the communication link, obtains accurate network state information, and correctly adjusts the data sending rate of the sending end, thereby improving the overall performance of the system.

Disclosure of Invention

The invention aims to provide a transmission control method in a wireless communication process, which can adapt to the current link state, distinguish link packet loss and congestion packet loss, and ensure that a communication system can ensure network performances such as high throughput, stable transmission and the like under different network conditions.

The invention provides a communication method for self-adaptive adjustment based on a satellite network state, which has the following basic principle:

the conventional TFRC protocol is based on the TCPReno throughput equation (1) to calculate the data transmission rate X.

X is the transmission rate of the data stream;

s is the packet size;

r is the packet round trip time RTT;

b is the number of packets acknowledged in the ack response, usually taken to be 1;

p is the loss event rate, an important discussion variable in this document, which is calculated in the manner given below;

t_RTOis the timeout retransmission latency, which is 4 × R.

The main determining factor is the loss event rate p in the network, the loss event rate p is the reciprocal of the average packet loss event interval within a certain time, and the packet loss interval refers to the number of data packets between two packet loss events. Network congestion in a wired network is a main cause of data packet loss in a link, and the loss event rate p can reflect the current network state; however, in a wireless network, due to the existence of a certain wireless error rate, a protocol cannot distinguish packet loss types, all packet losses are determined to be caused by network congestion, the calculation of the loss event rate is influenced, the data sending rate is excessively reduced, and the performance of the system is reduced. Packet loss intervals of the wired network and the wireless network are respectively shown in fig. 1 and fig. 2, wherein numbers represent sequence numbers of data packets, a letter C represents congestion packet loss, and W represents wireless error code packet loss. Due to the existence of wireless error packet loss, the packet loss interval is reduced from 20(23-3) to 18(21-3), the loss event rate is increased, the transmission rate is reduced, and the network bandwidth utilization rate is reduced.

The carrier-to-noise ratio of the satellite network receiving end, namely the ratio of the carrier power C to the noise power N, can be obtained by calculation through a formula (2)

The EIRP transmitter is equivalent to an omni-directional radiated power.

L_tIs the transmit feeder loss.

L_rIs the receive feeder loss.

k is Boltzmann constant, k is 1.38 × 10^-23J/K。

B is the bandwidth.

T is the noise temperature of the noise source.

G/T is the figure of merit of the receiving system.

In the formula (2), the equivalent isotropic radiation power EIRP of the transmitter and the gain G of the receiving antenna are fixed values, and the free space transmission loss L_fIs calculated by the formula (3)

Since the relation between the operating wavelength and the frequency is λ ═ c/f, equation (4) is obtained by simplification.

The pi and c constants are substituted and simplified into formula (5), so that the free space transmission loss is mainly related to the link length d and the wavelength f.

L_f＝92.44+20lgd+20lgf (5)

L in the formula (2)_tIs the transmission feeder loss, L_rIs the receive feeder loss, k is the boltzmann constant,

k＝1.38×10^-23J/K, B is bandwidth medium orbit satellite generally 36MHz, T is noise temperature of noise source, G/T is quality factor of receiving system is inherent property of satellite, let L be L_fL_tL_rAnd it is defined that the link and system losses and margins are then:

C/N(dB)＝EIRP(dBw)+G/T(dBk)-L(dB)-k(dBw/k)-B(dBHz) (6)

digital satellite communication systems generally have two modulation and demodulation modes, BPSK and QPSK. The bit error rate formula is:

and the loss probability packet _ loss of a packet with size packet _ size is:

packet_loss＝1-(1-P_e)^packet_size (8)

thus the link error rate P_eThe estimation calculation can be carried out through the physical environment state of the satellite network, and the estimation calculation is mainly determined by the link length and the wavelength except the inherent properties and fixed values such as the quality factor of a receiver, the transmitting power, the bandwidth and the like. New packet loss interval C_i' calculated from equation (8), and finally substituted into equation (10) to derive the exact loss event rate.

C_i′＝C_i×(1+packet_loss) (9)

C_iIs a packet loss interval;

packet _ loss is packet loss rate;

w_iis a weight coefficient;

finally, the data is substituted into the formula (1) again to obtain the correct data sending rate.

The technical scheme of the invention is as follows:

1) the system receives a communication application of a user, and starts communication after the system agrees to the request;

2) judging the current network communication state by acquiring parameters such as communication satellite quality factor, communication bandwidth, user sending power, communication link length, transmission wavelength at this time and the like, substituting each parameter into a formula for calculation to obtain an initial sending rate, and completing operations such as communication connection and the like;

3) the system adaptively adjusts the loop to recalculate the loss each time k data packet transmissions are completedInter-packet space C_i' and generates a new data transmission rate X;

4) and finishing the communication.

Compared with the prior art, the technical scheme adopted by the invention has the following advantages: (1) the method can carry out self-adaptive adjustment according to the real-time state of the network, and main parameters in the equation do not need to be estimated and set, so that the dynamic change is convenient to adapt to the actual communication environment, and the method is more flexible; (2) fig. 3 is a simulation statistical chart of the bottleneck link throughput under different link error rate conditions in the method and the conventional method, and it can be clearly seen from the simulation statistical chart that compared with other transmission control protocols, the method has better adaptability and higher network efficiency.

Drawings

Fig. 1 is a schematic diagram of packet loss intervals in a wired network.

Fig. 2 is a schematic diagram of packet loss intervals in a wireless network.

Fig. 3 is a comparison graph of bottleneck link throughput of three transmission control protocols under different link error rates.

Fig. 4 is a schematic diagram of an actual communication scenario.

Detailed Description

The invention is further described below mainly in connection with fig. 4:

the experimental implementation process is as follows:

1. the use scenario of the method is shown in fig. 4, which includes a communication medium orbit satellite MEO 1, aclient 2 and aserver 3. The transmission control protocol provided by the method is deployed on an MEO satellite;

2. a client sends a connection request with a server to an MEO satellite, and request information comprises information such as a requested task type, a communication link distance between the client and the satellite, a transmission wavelength adopted by the communication and the like; after the satellite agrees to the client side to apply for permission, the state of a communication link between the client side and the satellite is judged according to the received information, and the transmission rate obtained by the algorithm is fed back to a sending end (client side). Meanwhile, the satellite can confirm information such as the link length between the satellite and the server and calculate the sending rate of the satellite end;

3. the method has dynamic self-adaptability: the protocol algorithm can reevaluate the network environment once every k data packets are transmitted by the transmitting end, particularly remeasure the length of a transmission link between the transmitting end and the receiving end, and calculate a new transmitting rate until the communication process is finished;

4. and (4) analyzing results: the experiment is mainly used for evaluating the adjusting and controlling performance of the method in a bottleneck link generated by satellite-ground wireless link connection. The simulation experiment result is shown in fig. 3, both TCP and TFRC in the diagram are traditional protocol algorithms, TFRC-EA is the protocol algorithm proposed by the present invention, and it can be clearly seen through a comparison diagram that as the link error rate increases, the throughput of TCP flow, TFRC flow and TFRC-EA flow all presents a downward trend, but different from the significant decrease of TCP and TFRC, TFRC-EA distinguishes link error and network congestion, so that the network state is well adapted to, the system throughput is maintained at a relatively high level, the data sending rate is guaranteed not to be affected, and the network utilization rate is improved, therefore, the method is feasible.

The above is a typical application of the present invention, and the application of the present invention is not limited thereto.

Claims (1)

Translated fromChinese

1.一种基于卫星网络状态进行自适应调节的通信方法，其特征在于：1. a kind of communication method that carries out self-adaptive adjustment based on satellite network state, it is characterized in that:本方法的实现过程如下，The implementation process of this method is as follows:1)系统收到用户的通信申请，系统同意请求后开始通信；1) The system receives the communication application from the user, and the system starts communication after agreeing to the request;2)通过获取通信卫星品质因数、通信带宽和用户发送功率以及通信链路长度和此次的传输波长参数判断当前网络通信状态，将各个参数带入公式中进行计算，得到初始发送速率，并以此完成通信连接操作；2) Judging the current network communication state by obtaining the quality factor of the communication satellite, the communication bandwidth, the transmission power of the user, the length of the communication link and the transmission wavelength parameter of this time, and bringing each parameter into the formula for calculation to obtain the initial transmission rate, and using This completes the communication connection operation;3)系统自适应调整环节，每完成k次数据包的传输，就重新计算丢包间隔C_i'并生成新的数据发送率X；3) system self-adaptive adjustment link, every complete transmission of k data packets, just recalculate packet loss interval C_i ' and generate new data transmission rate X;4)本次通信结束；4) This communication ends;TFRC协议是基于TCP Reno吞吐量方程式(1)计算得出数据发送速率X；The TFRC protocol is based on the TCP Reno throughput equation (1) to calculate the data transmission rate X;

X是数据流的传输速率；X is the transmission rate of the data stream;S是数据包大小；S is the packet size;R是数据包往返时间RTT；R is the packet round trip time RTT;b是ack应答中确认的数据包数量，通常取1；b is the number of packets confirmed in the ack response, usually 1;p是丢失事件率；p is the lost event rate;t_RTO是超时重传等待时间，为4×R；t_RTO is the timeout retransmission waiting time, which is 4×R;决定因素是网络中的丢失事件率p，该丢失事件率p为一定时间内平均丢包事件间隔的倒数，丢包间隔指的是两次丢包事件之间的数据包数量；在有线网络中网络拥塞是造成链路中数据包丢失的主要原因，丢失事件率p能够反应当前的网络状态；但在无线网络中由于存在一定的无线误码率，协议无法区分丢包类型，将所有丢包全部判定为网络拥塞造成，丢失事件率的计算受到影响，过多的降低数据发送速率，造成系统性能下降；有线网络和无线网络的丢包间隔中，字母C代表拥塞丢包，W代表无线误码丢包；The determining factor is the loss event rate p in the network, which is the reciprocal of the average packet loss event interval within a certain period of time, and the packet loss interval refers to the number of packets between two packet loss events; in a wired network Network congestion is the main reason for the loss of data packets in the link, and the loss event rate p can reflect the current network status; but in a wireless network, due to a certain wireless bit error rate, the protocol cannot distinguish the type of packet loss, and all lost packets All are determined to be caused by network congestion, the calculation of the loss event rate is affected, and the data transmission rate is reduced too much, resulting in system performance degradation; in the packet loss interval of wired and wireless networks, the letter C represents congestion loss, and W represents wireless error. code packet loss;卫星网络接收端载噪比即载波功率C与噪声功率N之比通过公式(2)计算得出The carrier-to-noise ratio at the receiving end of the satellite network, that is, the ratio of the carrier power C to the noise power N, is calculated by formula (2)

EIRP发射机等效全向辐射功率；EIRP transmitter equivalent isotropic radiated power;L_t是发送馈线损耗；L_t is the transmission feeder loss;L_r是接收馈线损耗；L_r is the receive feeder loss;k是玻尔兹曼常数，k＝1.38×10^-23J/K；k is the Boltzmann constant, k=1.38×10^-23 J/K;B是带宽；B is the bandwidth;T是噪声源的噪声温度；T is the noise temperature of the noise source;G/T是接收系统的品质因数；G/T is the quality factor of the receiving system;式(2)中，发射机等效全向辐射功率EIRP和接收天线增益G是固定值，而自由空间传输损耗L_f通过公式(3)计算In Equation (2), the transmitter equivalent isotropic radiated power EIRP and the receiving antenna gain G are fixed values, and the free space transmission loss L_f is calculated by Equation (3)

由于工作波长与频率的关系为λ＝c/f，所以化简得到公式(4)；Since the relationship between the working wavelength and the frequency is λ=c/f, formula (4) is obtained by simplification;

将π、c常数带入，简化为公式(5)，由此看出自由空间传输损耗主要与链路长度d和波长f有关；The π and c constants are brought in, and simplified to formula (5), it can be seen that the free space transmission loss is mainly related to the link length d and wavelength f;L_f＝92.44+20lgd+20lgf (5)L_f =92.44+20lgd+20lgf (5)令L＝L_fL_tL_r，并将其定义为链路及系统损耗和余量，于是有：Let L=L_f L_t L_r and define it as link and system loss and margin, then we have:C/N(dB)＝EIRP(dBw)+G/T(dBk)-L(dB)-k(dBw/k)-B(dBHz) (6)C/N(dB)=EIRP(dBw)+G/T(dBk)-L(dB)-k(dBw/k)-B(dBHz) (6)数字卫星通信系统通常有BPSK和QPSK两种调制解调方式；其误码率公式为：Digital satellite communication systems usually have two modulation and demodulation methods, BPSK and QPSK; the bit error rate formula is:

而大小为packet_size的数据包的丢失概率packet_loss为：And the loss probability packet_loss of a packet of size packet_size is:packet_loss＝1-(1-P_e)^packet_size (8)packet_loss=1-(1-P_e )^packet_size (8)于是链路误码率P_e就能够通过卫星网络的网络通信状态进行预估计算；新的丢包间隔C_i'由公式(9)计算得出，最后带入公式(10)得出准确的丢失事件率；Then the link error rate_Pe can be estimated and calculated through the network communication state of the satellite network; the new packet loss interval C_i ' is calculated by formula (9), and finally brought into formula (10) to obtain an accurate Lost event rate;C_i′＝C_i×(1+packet_loss) (9)C_i ′=C_i ×(1+packet_loss) (9)

C_i是丢包间隔；C_i is the packet loss interval;packet_loss是丢包率；packet_loss is the packet loss rate;w_i是权重系数；w_i is the weight coefficient;最后重新带入公式(1)中，得出数据发送速率。Finally, it is brought back into formula (1) to obtain the data transmission rate.

Images